Monitor mode to live view on tv or monitor

[Victoria Wang]

I’d like to see if camect could have a monitor mode, that it will allow users to view all the cameras together on a tv or monitor while connect it with a HDMI cable.

[Will Stillwell]

Something to think about with this idea however is the overhead that puts on the Camect.  Currently the Camect doesn't have a presentation layer.  That runs on your device viewing the cameras.   It would require the camect to actually have some form of GUI.  Diverting cpu to handle that rather than processing video is not a good idea.   You then have to consider keyboard and mouse control with it.   Personally I need as much of the Camect to work on video processing, and not on presentation so much.  Leave that rendering to the devices that displays.  If you really want it on a monitor then use some other device connected to the monitor to do the presentation layer.  

~Will

On Sun, Aug 9, 2020 at 12:45 AM Victoria Wang <yafe...@gmail.com> wrote:

I’d like to see if camect could have a monitor mode, that it will allow users to view all the cameras together on a tv or monitor while connect it with a HDMI cable.

--
You received this message because you are subscribed to the Google Groups "Camect User Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to forum+un...@camect.com.
To view this discussion on the web visit https://groups.google.com/a/camect.com/d/msgid/forum/7871c7e9-8ada-455e-9464-f1c9026bbb6do%40camect.com.

[CamectArup]

As Will said, doing local display via HDMI would take up quite a bit of resources, so we're unlikely to implement that. What we'd like to do eventually is add the ability to send video to a cast receiver (e.g. Chromecast, or any Android TV) as a way for you to monitor cameras. 

We're also aware of a user who put together his own monitoring setup by sideloading chrome onto a firestick, and plugging the firestick into his TV. We haven't tried this ourselves so I don't have any first-hand knowledge of how well it works. 

[Brian Robertson]

I can say that we have Android TV boxes on our TV’s that we use for streaming. We have Google Chrome installed and run Camect on it all the time to watch our cameras. The wife loves it because one of our cameras in for the nursery and she can pull up that camera on the TV to watch our son while she is doing stuff around the house.

[Will Stillwell]

This is a great idea.  I'm thinking I might need to see if I can get it up and going on my fire tv stick.     Is it possible to create a browser based shortcut link that goes straight to a specific camera?  Hmm    

~Will

On Sun, Aug 9, 2020 at 5:12 PM Brian Robertson <brianmr...@gmail.com> wrote:

I can say that we have Android TV boxes on our TV’s that we use for streaming. We have Google Chrome installed and run Camect on it all the time to watch our cameras. The wife loves it because one of our cameras in for the nursery and she can pull up that camera on the TV to watch our son while she is doing stuff around the house.

--
You received this message because you are subscribed to the Google Groups "Camect User Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to forum+un...@camect.com.

To view this discussion on the web visit https://groups.google.com/a/camect.com/d/msgid/forum/CE783091-C777-42C7-BFD7-A024B8EAFDA5%40gmail.com.

[Jack 7]

Another method is to add the IP Camera Viewer channel to a Roku. It provides full screen viewing of one camera at a time, or four cameras in Split Screen View.

[Victoria Wang]

Thank you Arup and Will.

[Victoria Wang]

Thanks for the info! Could you please tell more details on how to run it on TV? Just purchase a box like Roku, then install chrome on it and log into the camect?

[Eric Meeson]

For curiosity, why is local display on HDMI more resource intensive than streaming it to another machine to display? Is it just because you're already using GPU processing to accelerate the detection?

[CamectArup]

Because you'd have to replicate all of the stuff the browser normally does that Camect currently does not ... i.e. render the UI, handle interactions with the user, decompress and play the video, etc. Imagine how much resources a browser requires to do this, and expect that it would take something comparable to do it inside Camect to show the full UI. A dumber display (e.g. just show video, with no other UI) might take somewhat less ... but in terms of where we'd prefer to put limited development resources, a cast-based solution (e.g. send to chromecast or an Android tv) would be useful to a lot more people as it would not require Camect to be located somewhere physically close to the tv. 

[Will Stillwell]

I would MUCH rather have the camect resources dedicated to capturing and recording cameras.   I'm to the point that I'd want a more powerful camect as it is, or build my own hardware.  It can borderline handle my 10 cameras.  And I want to replace 3 or 4 ring cameras with POE 4MP plus cameras.   I'd bring that 10 to 14 and there is no way Camect can handle what I want with how it is.  If resources were diverted to HDMI output with the UI that would be even worse. 

~Will

On Sun, Aug 16, 2020 at 3:10 PM CamectArup <ar...@camect.com> wrote:

Because you'd have to replicate all of the stuff the browser normally does that Camect currently does not ... i.e. render the UI, handle interactions with the user, decompress and play the video, etc. Imagine how much resources a browser requires to do this, and expect that it would take something comparable to do it inside Camect to show the full UI. A dumber display (e.g. just show video, with no other UI) might take somewhat less ... but in terms of where we'd prefer to put limited development resources, a cast-based solution (e.g. send to chromecast or an Android tv) would be useful to a lot more people as it would not require Camect to be located somewhere physically close to the tv. 

--
You received this message because you are subscribed to the Google Groups "Camect User Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to forum+un...@camect.com.

To view this discussion on the web visit https://groups.google.com/a/camect.com/d/msgid/forum/f95010b5-a743-4aa4-86bf-05f607b5c1e0o%40camect.com.

[MOH]

Albeit as a new user, I agree 100% that this is not where I would want resource allocation. The major limitation with the current HW is the priority placed on small form factor over power (I’m sure that careful consideration and research was done to determine the target customer, their needs in terms of the average number of cameras/household, and sweet spot regarding pricing resulting in the current choice of HW). I am very impressed with the system (am very amazed as how good the object recognition is given the very limited cpu and give the development team due respect for achieving this), but I purchased the system with reservations due to the limited load the cpu could handle. Am currently only using Camect as a notification device for object recognition to minimize false positive alerts due to the limited MP it can handle. Adding more cpu load for video display would be a HUGE mistake IMHO. What we need is a more scalable system facilitated by official HW that can receive support (so Camect HW). Of course the Camect’s business model must survive and so I support them and understand that as a startup they must also address more pressing/issues and high demand features. Their responsiveness and technical support is second to none and this played a large role in my decision to take a chance with this system, as limited as the HW currently is for my needs (multiple 4K cameras).  Not  only is the cpu limited, but as a business, company resources are also limited as well, and at this time, I would not squander either on a monitor function. Perhaps if the market shows that a substantially more powerful model would make sense from a revenue perspective can these cpu intensive features be incorporated. Just MHO as a recent user.

[R Spivack]

Another way to approach capacity/scalability would be a virtual grouping feature to combine multiple Camect boxes into a single GUI presentation.

That would allow horizontal scaling by adding more Camect boxes but being able to manage them from a single web browser and see a matrix view of all cameras without regard to physical box boundaries.

Of course, a trade-off in design and resources, but has the advantage of modular scalability, re-using a single hardware platform versus supporting multiple hw boxes of different power/performance.

The approach of offering self-builds of custom configured platforms has it's own larger set of trade-offs and commercial viability.

Haven't thought this through myself, so can't say which approach I would favor, just surfacing a different way to achieve the goal of higher density/higher camera capacity.

[Michael H]

Modular scalability sounds like a great approach. Advantages to Camect would be simpler inventory/manufacturing, and simpler HW support and tech training due to less variation in HW models. I guess the viability would depend on current margins on the current HW vs. subscription margins vs. cost of support and infrastructure. On the consumer end, for me at least, not losing the utility of existing units while adding scalability would definitely be a plus and would warrant a premium for this feature over other systems. The current form factor also is well suited to this as they are compact and the relatively low power cpu’s are less prone to over-heating. Units can even be decentralized in the home for security reasons (I know thats a little bit of a stretch but after all thats why DARPA decentralized the net, lol!). Anyway, add my vote to this as a desired feature.

[Will Stillwell]

Agree, group multiple units into 1 UI would be great.   View multiple camect feeds into 1 UI.   Have 1 master said of rules / notifications so forth.  So if you 'Snooze" your cameras it does it across multiple camects.   This would be the best way to address high MP feeds.   3 8MP feeds per unit.   You could easily do 3 or 4 units in one UI grid wise.   But I'll say this.  The cost of this vs a single Core i5 4 core or more core CPU I'm not sure it would be cost effective.  BUT each unit would have it's own 1TB storage vs having a single unit having to have a bigger single HD.   Everything generating more heat.   

Arup is this anything we are likely to see?  Higher Horsepower Single box, or multiple boxes under 1 UI?

~Will

To view this discussion on the web visit https://groups.google.com/a/camect.com/d/msgid/forum/CAJJeQJDycuzYcoUc3rjdEoWEtOhVvX2H8nae-FpYttqKMG0hww%40mail.gmail.com.

[Brian Robertson]

It would be nice to connect multiple boxes and/or have an option for a larger and higher performing box. I would definitely use that feature if it was there so I could have higher quality streams.  

On Aug 17, 2020, at 8:48 AM, Will Stillwell <will.st...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/a/camect.com/d/msgid/forum/CANM50hm5sJ6iUR%2Bhno79HbSw74o1Z00j00zhit2Vptb5Htw7dg%40mail.gmail.com.

[Michael H]

Regarding potentially higher cost per MP with multiple units to the consumer, as a consumer, I would be willing for a couple of reasons:

1. I’m willing to pay extra for flexibility and scalability.

2. This flexibility/scalability allows for lower entry costs. You don’t have to go all in if you want to try out the system (as I did) as a “test” to check out the superior AI.

3. Like #2, allows for lower entry cost for those that already have NVRs that may wish to use the device primarily as a lower res notification device for object recognition expanding the market for those who may not wish to completely replace an existing system already invested in.

4. Although overall cost per MP capacity may be higher for those wishing to completely replace a high capacity system (due to having to purchase multiple units), the ability to grow the system incrementally is a big plus in my mind and would justify the added cost IMHO (at least for me).

5. Redundancy is not the right term, but the ability to redistribute camera res among units as you grow a system (for example as you add units, you can up lower resolution cam streams to higher res streams on your more critical cameras, even if the total number of cameras don’t change) is again a big plus and worth an overall price premium.

6. Would definitely increase repeat sales to those already familiar with and happy with the system but wanting HW support for a higher capacity system without effectively bricking a previously purchased unit by replacement with a non-modular higher capacity unit.

7. I’m certainly no expert on this but it seems to me that this would be primarily a software implementation that may be implemented through the web interface. Any additional cpu load may perhaps be handled by one of the units which could designated as a hub or the master unit (this unit may not be able handle a full 24MP camera load but that's understandable).

Enough reasons, lol?  Arup, please consider this! There must be many other users that would love this capability. I understand that priority must be given to features that have the most impact on expanding market potential and retention (i.e. expanding user base, addressing bugs, etc) but I am essentially happy with the functionality of the system as is except for scalability limitations. This thread may be more appropriate in the Requested Features Discussion, please move if appropriate. On another note, a polling/voting system for requested features would be fantastic in the forum.

Thanks!

[Will Stillwell]

I can agree 100% with all 7 thoughts Michael.   

~Will

[Eric Meeson]

I had never paid attention to resource usage of decompressing the streamed video. That makes perfect sense when stopping to think on it. I just looked in Chrome's task manager and see that the small view for my 10 cameras pulls around 20% CPU usage! Obviously not good to try to do on the box itself.

I'll have to look at getting one of those sub $30 Android TV boxes to stream the video locally. Probably with a burner account rather than the one I use for control though, because the security of those low cost Android boxes is always a bit suspect. That or just use the camera streams directly as Jack mentioned and not load the Camect even with the streaming since it is for a local display anyway.

As a thought for future development, since the Android TV boxes can be had so inexpensively anyway, what about turning out a customized dedicated viewer box? This could let you have the Android TV get all the camera information (RTSP links and passwords) from the Camect itself, but handle all the streaming directly, only pulling data from the Camect for replay. This way live view wouldn't tie up any resources at all, and someone using it for say a business could have multiple viewers scattered around without worrying about total streaming load on the Camect itself.